Snow identification

Project that predicts whether a patch of land is snow free or not

The method is based on the mean grayscaled pixel value around the center of the image. We take a box of (by default 25px by 25px) around the center point of the image, predict the mean grayscaled pixel value and compare it to a threshold.

If the value is lower than 0.33, we consider the image to be snow free.

If the value is from 0.33 to 0.66, we consider the image to be partially snow covered.

If the value is higher than 0.66, we consider the image to be snow covered.

Prerequisites

To start the project, we need to create a table in the sqlite3 database.

To do so, run the following command:

python -m init_db

Virtual env

To create the virtual env using anaconda env.yml file, run the following command:

conda env create -f env.yml

To update the env, use the command:

conda env update -f env.yml

Project structure

All the images should be put into the input/ directory.

The output/ directory will contain a json with the image name and the result of the prediction.